Finetuning
Query finetuned models
You can start querying the finetuned model by replacing the model name in your existing OpenAI code with the finetuned model. Here is a simple example:
- Get the ID of the finetuned model from the dashboard. The highlighted part in the clip below is the MODEL_NAME_FROMCONSOLE.
- Use it in your existing code:
You start using your finetuned models by changing the API end point
Make sure that you installed gigaml sdk
python
pip install gigaml
Here is how you can start using the finetuned model
python
import openai, os
from dotenv import load_dotenv
# IMPORTING GIGA ML LIBRARY
from gigaml import OpenAI, GigaMlApi
load_dotenv('.env')
client = OpenAI(
api_key = os.environ.get('OPENAI_API_KEY'),
#Initialise gigaml Client through API_KEY which can be generated via our console
gigaml_client = GigaMlApi(token = os.environ.get("GIGAML_KEY")),
)
response = client.chat.completions.create(
model = MODEL_NAME_FROMCONSOLE,
messages = [{"role":"user", "content": "Who are you"}],
tags = ["prod"],
stream = True,
)
for chunk in response:
print(chunk)
Note that, for querying LoRA models, there won’t be any cold start delay. If you are querying a full-finetune model, there will be a cold start delay of 10-15 seconds.